The Great AI Debate

As the great wave of AI has transformed the world, emerging automated technology has entered more and more into the fabric of our lives. From personal virtual assistants and chatbots to self-driving vehicles and telerobotics, AI has become threaded into large tracts of everyday life, and increasingly reshaped the economy and society. Klaus Schwab (2016), founder of the World Economic Forum, has contended that AI ushers into existence a "fourth industrial revolution". The first profound industrial revolution was steam-powered, the second electrical, the third the arrival of the computer age, and Schwab argues that today's AI revolution is "unlike anything humankind has experienced before". AI is not so much an advancement of technology, but rather the metamorphosis of all technology.

But what, exactly, is "revolutionary" about AI? To begin with, AI profoundly changes politics. Technological change and how to achieve digital literacy across society, it might be argued, are now the name of the game in key areas of politics. The automation of jobs has become the most debated of them all, but others include self-driving cars, sex robots, socially-assistive robotics, killer drones, and many other areas (Bissell et al., 2020; Boyd, 2024; Döring & Pöschl, 2018; Elliott, 2019; Endo, 2020; Guess & Munger, 2023).

In many respects, it is a new agenda today, at least in terms of public policy. The traditional welfare state, for example, was very much based upon the provision of remedies to the fallout from problems once they had happened - if you lose your job, the state will provide benefits until you find a new one. Today, we live in a very different world. We live in a time where robots move boxes in factories as well as conduct shelf-auditing in supermarkets, and where complex algorithms complete tax returns and trade on financial markets. The consequences of our increasingly automated global world involve a shattering of political orthodoxies. As a consequence, politicians and policy-makers need to be much more interventionist, crafting policy thinking to cope with the unexpected, unanticipated shifts stemming from the digital revolution.

Another major consequence of the AI revolution has been that our entire parameters of risk and opportunity radically change. Complex systems such as machine learning and advanced robotics are characterized by unpredictability, uncertainty, zigzags, and reversals. Policy attempts aimed at coping with complex technological systems often produce unintended consequences, revealing or generating other issues or problems, and other solutions or synergies in turn emerge in response to such interdependencies between adaptive systems. This has major ramifications for nation-states, which not only exist in systemic relations with other nation-states but have now to cope with the spread of artificial intelligence and its global offshoots, including offshoring, outsourcing, reshoring and the like.

The rise of AI has stirred massive global controversy. Not just in universities and think-tanks, but across industries and business circles. For many, the very idea of AI generates alarm. The advancement of AI represents a major threat to jobs, employment, enterprise, and industrial manufacture, stoking up anxieties that pervade many other areas of life too. The more powerful automated systems become, the more many people worry about the risk that AI might result in some irreversible global nightmare. In this scenario, the rise of AI is potentially catastrophic. AI is equated here with an apocalyptic social future¹. Another response, albeit very different in perspective, views these technological breakthroughs more positively. This response concentrates on new possibilities, hopes – and dreams for a better world. On the face of things, AI is a breakthrough science, promising great opportunities for the reshaping of economies, societies and political choices today. In this scenario, the coming AI revolution foreshadows the opening of a new era, one which will radically transform people's daily habits and the world in which they live. AI is a new driver of production and will generate new sources of economic growth, changing how work is done and dramatically increasing growth in businesses worldwide.

Contrasting hypotheses, interpretations and theories of AI in the social sciences cluster around two main lines of argumentation. The first of these I shall refer to as the position advanced by sceptics. The sceptics are today in a minority and yet have significantly influenced public opinion and much policy thinking on AI and its ramifications. Simply put, sceptics say that claims of an AI revolution are overblown. For many of a sceptical persuasion, the spectre of AI is too often invoked to explain away complex institutional changes occurring throughout the world today. These are changes to do with the international economy, workplace change and geopolitics (Bender & Hanna, 2025; Capello et al., 2022). The contrasting, second position is occupied by those I call transformationalists. The AI revolution, argue transformationalists, is creating a world of radical change. This is the dawn of a new era, one in which the intersecting forces of economy, society and politics shifts in fundamentally new directions. From this perspective, AI powerfully disrupts traditional ways of doing things, ushering in new economic conditions, social divisions and political alignments (Gruetzemacher & Whittlestone, 2022; Holmström, 2022).

Sceptics of AI characteristically focus upon the many and varied overblown claims of technological innovation and scientific advancement in the age of intelligent machines. Instead of providing insight into the technological forces reshaping our standards of living and possibilities for the future of humanity, the discourse of AI, argue many sceptics, obfuscates social, economic and political changes that are actually happening. Broadly speaking, sceptics say that the discourse of AI is used to explain away complex contemporary problems – especially those stemming from workplace change, international economic convergences and geopolitics. In summary, sceptics contend that the great AI revolution is no more than "hype".

Authors who write from this sceptical position believe that the impacts of AI are not dramatic and will take a considerable time, possibly many decades, to come to fruition. Should one pay any attention to claims of an AI revolution? Many sceptics say "no". What, ask the sceptics, is "revolutionary" about AI? Rejecting the equation of technology and transformation, this line of sceptical criticism develops a conception of workplace change which focuses squarely on the adaptability of employees, the enhancement of skills and new forms of economic efficiency and organizational innovation. Such a standpoint thus ties talent and technology firmly together. Sceptics stress the continued primacy of workplace productivity, workers' skills and emerging patterns of adaptation to technological innovation. From this standpoint, there is a clear disjuncture between the widespread innovations of AI and the world of work in which, for the most part, employees adjust to new technologies and find continued opportunities to acquire skills and capabilities.

By contrast, transformationalists reject the claim advanced by sceptics that AI is a synonym for hype, or a cover-story for tech companies. Whilst certain sensationalist aspects of the discourse of AI are not denied, the transformationalist response emphasizes that AI is an expression of deeper technological shifts in the scale of economic organization and social relations worldwide. This can be discerned, argue transformationalists, in the rise of advanced automation, supercomputers, 3-D printing, Industry 4.0 and the Internet of Things. AI technologies, including robotics and advanced digital systems that deploy deep learning, neural networks, machine decision-making and pattern recognition, have given rise to an era of intelligent machines which can increasingly sense their own environments, think, learn and react in response to data.

Transformationalists question the idea that economy and society can be adequately grasped from the business-as-usual perspective advanced by sceptics. The extensive penetration of the global economy by digital forces has fundamentally altered its operations and dynamics. Transformationalists generally underscore the essential significance of the digital revolution, an historic moment in the worldwide transformation of manufacture and services. This involves locating contemporary patterns of globalization within the new technological revolution, and a dazzling variety of terms have been coined to capture these momentous shifts - including "Industry 4.0", "digital capitalism", "algorithmic governmentality", "bot economy" and "automated society". In transforming both the conditions and consequences of the economy and society, according to this argument, AI has revolutionized corporate life and businesses across the world. Advances in machine learning algorithms and big data underpin extraordinary innovations in the manufacturing of goods and services as well as the emergence of new industries, and consequently, jobs and employment have come under assault as never before. The impact of smart machines, and of social media more generally, has significantly transformed the consumer economy itself. At the same time, these unparalleled technological innovations directly impact upon issues of ethics and governance. Recognizing how closely the impact of AI on jobs and public policy are intertwined, governments worldwide have sought to introduce a raft of measures geared towards enabling robust engagement with the digital revolution².

What should we make of these claims and counter claims? In my writings over the last ten or so years, I have demonstrated that AI impacts upon not only work, employment and unemployment, but also social relationships in the broadest sense. We can adequately grasp the social impacts of AI only if we question the general notion that digital technologies leave relations between people essentially unaltered. In The Culture of AI(2019), I argued that AI involves new forms of social interaction, new kinds of social relationships and new ways of experiencing and performing our identities. In conditions of accelerated AI, the influence of digitally mediated materials on social relationships, and on personal intimacies of the self, becomes more pronounced. Cloud computing, smart machines and predictive recommendation technologies play a central role in this respect. But so too do smart products, services and devices, 3D printing, intelligent eco-systems, virtual reality, augmented reality and supercomputers. In a fundamental way, this informational overlay (Thrift, 2011) reorients social life away from face-to-face interaction, which has traditionally characterized social organization, and generates new forms of digitally mediated interaction which transform our basic coordinates of time and space. The development of automated emerging technologies, powered by AI, significantly contributes to the reorganization of both self-organization and social interaction, up to and including the global digital economy.

The Theory of Algorithmic Modernity

What emerges from the great AI debate, among other things, is that discourse has shifted from the analysis of whether or not the advanced digital revolution is happening to consideration of the consequences of AI across economies, societies, polities and cultures. In a series of interconnected books – The Culture of AI (Elliott, 2019), Making Sense of AI (Elliott, 2021), Algorithmic Intimacy (Elliott, 2023) and Algorithms of Anxiety (Elliott, 2024) – I have argued that the development of AI has played a more fundamental role in transformations of modernity than many scholars in the social sciences have hitherto recognized. In these works, I set out to develop a new social theory of the advanced digital age that would do justice to the transformative character of AI. I formulated a particular theory about how the key technologies of the advanced digital age are 'rewiring' self and society - what I called algorithmic modernity. The basic idea of this theory is that the AI revolution is a here-and-now phenomenon which needs to be analysed in relation to the kinds of social action and cultural interaction which such technological innovations make possible in multiple affordances. The advanced digital age should be analysed not in the bombastic terms of techno-utopians ("everything changes because of AI"), but as bound up with the workings of the global economy and transformations of modernity. Mindful of the limitations of technological determinism, I have argued that algorithmic modernity significantly transforms advanced capitalism, lifestyle change, social inequalities and surveillance – and, in doing so, reshapes and twists emerging technologies to unexpected uses.

Algorithmic modernity signals a fundamental reordering of contemporary social life, from its institutions and structures to its cultural values, against the technological backcloth of real-time digital automation, datafication and the ubiquity of smart algorithms. In the era of predictive analytics, of quantified culture, a new kind of capitalism, a new kind of labour, a new kind of surveillance, a new kind of self-identity, indeed a new kind of society, begins to crystalize (Borch & Min, 2023; Burrell & Fourcade, 2021; Chioda, 2024; Zuboff, 2019). Unlike the early stages of digital modernity, which played out through the computational galaxy of the internet and collaborative global networks, algorithmic modernity unfolds across machine-learning systems, colossal 'big data', the predictive logics of recommendation technology and the increasing automation of lifestyle options. Especially since the 2010s – a time of exponential and risk-fraught technological innovation in AI, virtual and augmented realities, the Internet of Things, cryptocurrency and blockchain – society has transformed from a conveyer belt of individualization to singularization and culture from an engine of reflexivity into a generator of machine learning algorithms.

This technologically infused machine intelligence of the entire social field operates in terms of both extensity and intensity. At the level of extensity, what this refers to is the formation of trans-territorial complexes of automated digital activities, such as AI-based trading bots in financial spaces or data-mining and predictive risk modelling in welfare systems. The technological explosion of AI is vital to understanding the scale and scope of this new organizational paradigm. In the algorithmic version of modernity, automated emerging technologies are at the cutting edge of corporate profit and innovation; of the organizational manipulation and sale of data extracted from human experience to 'behavioural futures markets'; and the offshoring of socio-economic activities to cheap labour spots around the globe.

The late digital age is often spoken about as if it were simply a technological affair, but it is not. The digital revolution and the advances of artificial intelligence are deeply bound up with transformations of identity. This brings me to the theme of intensity. What have been the consequences of the digital revolution for the experience of the individual self? The psychology of the self in conditions of algorithmic modernity, I have argued previously at some length, is one requiring ongoing balance between hope and anxiety. Among the traits of this new subjectivity has been the powerful excitement, or reservoir of hope, that people can at last take control of a better future through the relegation of mundane activities to automated smart machines. Life lived on 'auto-pilot' promises novel opportunities to do other, more exhilarating, things. But, at the same time, people find that they must 'retrofit' what they see on their screens, from social media posts to 'personalized' recommendations, into the fabric of their lives. This phenomenon of always scrambling to 'keep up' helps create a psychic landscape infused with data fatigue, information overload, digital device distraction syndrome, and emotional burnout³.

Conventionally, burnout denotes an individual inability to cope with psychological stress. My argument is that fears, forebodings and anxieties over burnout become a general condition of algorithmic culture, reinforced as an infinite loop through the performativity of predictive analytics and recommendation technology. Automated social media, recommendation engines, predictive analytics, biometric tracking: the self in conditions of algorithmic modernity is increasingly recast in terms of calculations, metrics and quantification (Cheney-Lippold, 2017; Lupton, 2020). Fear of 'loss of identity' in this brave new world of automated smart machines is prone to leave psychic life depleted and diminished, marked by feelings of emptiness or boredom. Thus, the age of smart machines does not lessen social disquiet but exacerbates the anxieties it claims to repair, producing Algorithms of Anxiety, where personal identity is left shrunken and emotionally withdrawn.

Having captured the extensity of algorithmic modernity against the structural and geopolitical changes of the global economy, as well as the pervasive decentring of identities under the digital revolution, one core challenge for social theory today is to develop fresh thinking in a world confronting a new logic of change itself. The cultural logics at play here include the compression of time into the immediacy of real-time feeds, the stretching of space into virtual assemblages of users, devices, and systems, and the promotion of predictive analytics and behavioural nudging to maximize corporate profits (Mills, 2022; Rouvroy & Klumbytė, 2022). This ongoing transformation is producing a new configuration of modernity: a reorganized capitalism reliant on platforms, a redefinition of labour into task-based gig work and content moderation, a splintering of subjectivity into data profiles, and a hollowing out of traditional temporalities of work, care and governance. Individuals find themselves living beside their own digital doubles - that is, digital agents that sort, filter, and act on their behalf, often outside their awareness or control. Social life is increasingly coordinated not only through norms and institutions but through adaptive machine systems that evolve in tandem with the behaviour of social actors.

What we are witnessing, then, is not merely the deepening of individualization - as was the case in earlier forms of modernization - but an automization of the self. Individual preferences are no longer simply chosen; they are inferred, nudged, and sometimes fabricated through constant algorithmic tracking. Decision-making becomes less and less an act of personal deliberation and more a by-product of data flows managed elsewhere. The principal coordinates of algorithmic modernity consist of a new sociotechnical order where algorithms, apps and automation are deeply bound up with transformations in human agency, identity and society at large. Such transformations involve a wholesale shift from human-centered reflexivity to machine-driven sociality, with wide implications for subjectivity, power and ethics. The outline provided here of the theory of algorithmic modernity, an approach which to date has been necessarily partial and provisional, crosses and tangles in complex ways with other perspectives on the digital revolution developed by writers such as Bernard Stiegler, Helga Nowotny, Nigel Thrift, Shoshana Zuboff, Simona Chiodo and Sherry Turkle. (The most recent elaborations of the theory of algorithmic modernity have been developed in Elliott 2024 and Elliott, forthcoming).

Transformations of the Subject

But we need to press further. What are the consequences of algorithmic modernity for personal identity and selfhood? How might we rethink the constitution of subjectivity in the aftermath of the digital revolution? And what place to accord the theory of the subject in conditions of algorithmic modernization?

Life in algorithmic societies is increasingly predictive life. The world and all its living and non-living constituents are recast as objects of prediction: that is to say, objects whose value lies in their potential to be anticipated, calculated, and continuously reconstituted in advance of action. Self and society, from this angle, are remade after the model of computational predictive objects.

Some years ago, Apple in Japan ran an advertising campaign for the Apple Watch, one arguably illustrative of some of the key changes impacting personal life and self-identity today. The advert depicted people moving through urban space wearing the device strapped to their wrists whilst being nudged or prompted towards healthy, active pathways of movement. Entering social relations Apple-style involves movement not according to human intentions but rather through the incessant nudging of invisible anticipatory forces. The choreography of bodily movement, as women and men are perpetually pulled off intentional courses of action and redirected elsewhere, captures a world in which human subjects must constantly recalibrate their lives according to pre-emptive algorithmic recommendations and shifting digital terrains. A man walking towards his car is "pulled" by his Apple Watch and redirected towards a bicycle. A woman entering a building walks towards an office lift, but is suddenly "steered" by her device instead towards a staircase. A woman nearing the completion of a jogging session is "pushed" towards a stairway, only to continue further aerobic exercise. The advertisement presents a world where the individual subject is perpetually nudged, pulled, pushed, and steered. In this digital ecosystem, movement is not merely responsive but pre-emptive, not only adaptive but anticipatory, not purely self-referential but predictive.

This brings us directly to the transformed relations between self and society. We can discern a novel "push and pull" dynamic that bites deeply into social life today; this is a dynamic split between propensities towards anticipation inherent in the predictivity of complex digital systems on the one side, and disruptions of human agency on the other. Apple's Japan advertisement captures or exposes some key elements of this evolving cultural landscape where people come to feel outpaced by continual computational profiling, algorithmic predictions, and digital nudging of their everyday choices. It is not difficult to see this push-and-pull dynamic threaded throughout other domains of personal living. Every time someone receives an email notification chime when trying to complete a pressing task, casually looks at an incoming recommendation on their mobile device whilst talking with others, responds to system-level alerts ("install update now") while trying to find time for personal reflection, she or he tacitly acknowledges the "hijacking" or "rapid context-switching" which more and more defines life-conduct in algorithmic societies. It is, of course, the case that all sorts of counter-reactions to, as well as avoidance or rejection of, these prompts and nudges develop as individuals seek to cope with this monopolization of attention stemming from automated digital technologies. But more often than not, the taken-for-granted way in which individuals swerve or switch between everyday contexts of action and the intrusion of nonstop notifications, predictive recommendations, and constant task-switching testifies to the pervasiveness of technologies of distraction (Campo, 2022)

What is it like to live in a world increasingly saturated with emerging technologies of anticipation and prediction, namely predictive computational systems and their ever-tighter enmeshment with artificial intelligence? What kinds of subjects are produced in a world where datafication, metrics and quantification have come to the fore in the everyday lives of people? In an age of extensive algorithmic regimes and intensive predictive technologies, what happens to identity, personhood and the self across different forms of life - from the most intimate to the most public? Recent literature in science and technology studies, media studies and cultural theory tends towards dualism on these questions. Either human subjectivity and what we call personal identity is part of the medium of technology, or there is something about the self that is averse - something that stands apart - to the imprint of emerging technologies (Beer, 2022; Fisher, 2022; Hansen, 2015). In more everyday parlance, we could think of identity as technologically constructed on the algorithms go-all-the-way-down model; or we could think of personhood as in some oppositional sense external or beyond or excessive to emerging technologies.

These are, to be sure, caricatured portraits of the self in the digital age. But we should take seriously the complex ways in which such discourses have shaped the contrasting yet interlocking imaginaries surrounding advanced technology and the self. A faith in dynamism, metamorphosis, transformation, the power of endless self-reconstruction - all this profoundly permeates the individualistic, neoliberalism of Western culture as well as the strenuously self-affirmative ideology of Big Tech. That, crudely, emerging technologies, particularly transformational kinds of technology, underpin 'the good life' and 'the good society'. Underlying this rather upbeat brand of technologism lies an assumption that people are endlessly pliable, suggestible, flexible, capable indeed of perpetual plasticity. Indeed, with the rise of Silicon Valley and its high-tech satellites, technologism has today, in effect, gone global. It is no longer a question of the kinds of interaction between technology and society, as much as the spread of a dominant worldview that the latter is increasingly shaped by the former. The power of technological innovation, it is routinely assumed, will inevitably generate productive societal outcomes; and a certain kind of ensuing self-identity will ineluctably lead to certain kinds of lifestyle change, and the betterment of society.

At its starkest, then, it is as if we only have such alternatives as, say, social control under surveillance capitalism or digital detox programmes to break free from Big Tech (Stiegler, 2022). And either/or logic haunts contemporary studies of relations between technology and the self. Those authors who insist on the exquisite flexibility of self-reconstructing personhood, and for whom the 'dynamism' of technological development is unequivocally positive, tend to ride roughshod over the more culturally troubling or emotionally disturbing aspects of the digital revolution. In a similar vein, authors who posit technology's deterministic character from one end of cultural life to another tend to forget that people are a complex mix of personal convictions and emotional depths which are not easily amenable to the dictates of modern technology. But the relations between emerging technologies and selfhood do not need to hinge on such dichotomies. The self has not been 'deleted' by the profusion of automated smart machines, and the terminology of 'human-in-the-loop' does not sufficiently apprehend the quandary of the self in the digital age. There are other ways of understanding the intricate interconnections between selfhood and emerging technologies, and in what follows I briefly seek to outline this transformation in a different way.

The mediation of human experience through communication media and digital technologies has become deeply written into scripts of life. In the social sciences and humanities, there has been a strong focus in recent years on the nature of the self as transformed by the profusion of digital technologies. Broadly speaking, these lines of thought emphasize the unfolding of the time-space paths of daily life in which forms of personal and cultural experience become 'overwritten' or 'overdetermined' by mediated communication and digital technologies. The Internet, email, multi-user virtual worlds, social media: these were some of the computational interfaces in the early stages of the digital revolution through which human experience was 'stretched' and 'expanded' beyond the practical contexts of daily life. One of the most characteristic, and most discussed, changes that took place in this transformation of different kinds of experience during the early phases of the digital revolution was the rise of the 'online world', which was counterposed to the 'offline world'. In one picture of the self, the individual primarily moved through the time-space paths of immediate, physical proximity to others and the social world. In another picture of the self, digital experience became increasingly central to lived experience. Against this backdrop, various urgent questions came to the fore. To what extent should we incorporate the online world in a fuller picture of ourselves? What is it about the online world that makes it so difficult to negotiate when readjusting to the practical demands of everyday life? In what ways does digital experience at once support and sabotage the immediacy of experience in everyday life?

Reflecting on "the consequences of the online/offline split of the Lebenswelt", Zygmunt Bauman wrote:

The two universes differ sharply - in the worldview they inspire, the skills they require and the behavioural code ... they conjure up, patch together and promote. The differences can be, and indeed are, negotiated - but hardly reconciled. It is left to every person commuting between those two universes (and that means to all or almost all, and each or almost each, of us) to resolve the clashes between them and draw the boundaries of applicability of each one of the two disparate and often mutually contradictory codes/liturgies. But the experience derived from one universe cannot but affect the way in which we view the other universe, evaluate it and move through it; there tends to be a constant, legal or illegal but always heavy, border traffic between the two universes. (Bauman & Raud, 2015; pp 76-77)

What Bauman appears to be saying sounds like a familiar point: in our incurably diversified and technologically complex world, individuals confront daily demands to move between the world of digital technologies and the practical context of their everyday lives. But he is also saying more than this. He is saying that these contrasting worlds are inextricably entangled; he's saying that these worlds are full of numerous complexities and complications; and he is saying that there can be significant 'bleed' from the world of digital technologies to the practical engagements of the self.

But when Bauman wrote this in 2015, we should remember he was writing largely about the 'connected selves' of social media culture. At this time, Bauman made it abundantly clear that he did not think people were doomed because of social media and the Internet (and indeed warned of the dangers of technological determinism); but he did underscore that people's lives were increasingly spent managing the endless criss-crossings of these boundaries. As Bauman wrote: "Having a life divided between two universes, the online and offline worlds each with its distinct possibilities and requirements, and bound as we are to travel daily if not hourly to and from between them, we cannot avoid repeatedly bumping into a clash between the intentions and expectations brought from one universe, and the reality specific to the other: particularly if the first have been acquired, trained and entrenched online, while the second were confronted upon return to the offline domain". (ibid; 88)

These emphases, I think, sound right, but the sociological portrait already appears dated. Bauman's account that people are always 'commuting' between online and offline worlds has surely been outstripped - supplemented, expanded, redoubled - by developments in emerging technologies and innovations in artificial intelligence. Most importantly, what was called the 'online world' has metamorphosized from screens to ecosystems. As an ambient, invisible force field, AI nudges behaviour through unseen prompts, predicts actions based on past behaviours, as well as silently senses movement and adjusts environments before deliberative social activity is undertaken or performed. Screens, tablets and mobile devices remain at the centre of this traffic between self and society. But the 'online world' has become an 'algorithmic world' - one in which the unseen, the invisible and the indiscernible play an ever-greater role in reshaping self-formation as well as social reproduction.

Living in an automated world involves a continual 'back-and-forth' between the immediacy of everyday experience and the predictive intrusion of emerging technologies. While lived experience and physical proximity to other people remain at the core of the individual subject's encounters with self, others and the wider world, today's structuration of human experience is increasingly intertwined with automated technologies that anticipate personal needs, organise consumer choices and shape possible futures. This interplay between the practical contexts of daily life and the profusion of algorithmic prediction creates novel possibilities for self-expression, personal decision-making and social relationships, yet it also introduces new uncertainties and risks. Individuals find themselves having to negotiate between their own biographies and situated social contexts on the one side, and the calculations, recommendations and forecasts generated by automated intelligent systems on the other. As predictive technologies become more and more embedded within the routines of daily life, people are called not simply to adapt to automation but to move between social contexts of action and machine-generated anticipation – that is, between what is lived in the present and what is predicted about the future. The automated world thus expands the possibilities of everyday life while simultaneously requiring an ongoing reflexive balancing between personal agency and smart algorithms⁴.

At this point, we are now in a position to bring the foregoing elements of the analysis together. The contemporary self, I am suggesting, is increasingly formed through an ongoing switching or swerving between engagement with contextual settings of social action and encounters with algorithmic systems of calculation and prediction that increasingly transform how that world is experienced and represented. The increasingly automated features of the digital universe help to explain why the experience of the contemporary self is akin to that of a relentless lurching rather than that of connection or control. This is a psychic landscape whose experiential contours intersect with invisible, high-velocity algorithmic power wielded by large tech companies. In the age of big data and behavioural futures markets, this perpetual lurching suffuses the self with illusions of smooth, frictionless, predictive pathways of action on the one hand, whilst veering away erratically in directions mandated by proprietary black boxes that individuals often fail to decipher or foresee on the other. One consequence of this is a new distractedness of the individual subject, no longer anchored within secure interactional frames, where boundaries of self and other were routinely reaffirmed. In conditions of algorithmic modernity, by contrast, psychic life becomes increasingly episodic and disturbingly erratic, stamped by sudden shifts of attention or erratic frames of mind. This swerving, spasmodic flux of psychic life is, in many respects, the flipside of 'identities' disaggregated into various types of datum. Significantly, in the split between the self as biographical person and the self as data traces, the subject is rendered 'doubly decentred', caught in a paradoxical interplay of computational prediction and agentic performance.

Various contradictions are at work in this structural lurching itself. The more people become seduced by the alluring promise of instant algorithmic recommendation and effortless automation, the more their inner resources become depleted, thus exposing the self to considerable ambivalent tension. Selfhood, in effect, becomes deeply inscribed within emerging technologies which are constantly capable of submerging it in automated machine intelligence. The contemporary self is at once constituted and transformed through swerving across platforms, notifications and automated nudges, as daily life unfolds against a backdrop of novel threats and risks. These dangers – systemic bias, automated exclusion and pervasive surveillance – can appear in a variety of shapes, with a variety of patterns impacting both self-constitution and social reproduction.

It is difficult, however, to conceptually nail down such lurching without defusing it or rendering it invariant. To say this is to emphasize, once again, that selfhood after the digital revolution is not all of one piece, and here my account will require further refinement. The sudden tilts of movement which I have suggested increasingly define experiences of the self are not, of course, the same thing as uniform levels of subjective response to the global digital order. It should be clear that, within existing forms of human-machine interaction, algorithmic depersonalization (or decentring of the subject) is not universally determined. To characterize selfhood, identity and subjectivity in terms of the lurching bound up with digital technology engagement and usage involves a series of shifts marked by sporadic, contradictory 'pushes and pulls' – automated feeds pulling the subject's attention one way, predictive algorithms nudging personal choices another, as well as the fractured, switching between digital worlds and practical contexts of action which continually reassemble relations between self and others. Future attempts to capture the lived experience of this automated, algorithmic condition might well begin from this viewpoint, grounded in a careful appraisal of how contemporary culture draws upon and represses the inner compulsions of social actors.

Acknowledgements:

This paper was written during a 1-month residence at the Paris Institute for Advanced Study under the "Paris IAS Ideas" program. Some of the ideas developed here overlap with arguments I have set out in the following publications: Elliott, A. (2021). Making Sense of AI: Our Algorithmic World. Cambridge: Polity. Elliott, A. (2024). Algorithms of Anxiety: Fear in the Digital Age. Cambridge: Polity. Elliott, A. (2026); "Artificial Intelligence in the Aftermath of Algorithmic Modernity: Some Key Contributions and Open Issues in Japanese Sociology", Japanese Journal of Sociology, 35:e70004.